Klaas Waslander Blog

This blog post is about a pitfall that can happen with any process and tool. The fact is that it can happen more easily with processes and tools that people feel strongly about.

Scrum is a process that, if implemented correctly, has great benefits. To name a few: dedicated project teams (they commit) giving high innovation speed, transparent progress and control by releasing every sprint, close involvement of the customer in projects, self learning effect of teams.

On the other hand it can easily happen that sprints are followed so rigorously, that during a sprint period no daily tasks are picked up by team members anymore. This needs to be managed carefully, because it can lead to a feeling of inflexibility and “slowness” towards customers and stakeholders, while that is not necessarily the case. At the same time it can be easily fixed in various ways:

• Introduce shorter sprints allowing for a quicker response
• Take some flexibility into account in a sprint, allowing for daily tasks without immediate effect on the sprint planning (in Scrum terms: you can lower the focus factor)

However, even more important is what is “between the ears” of every team member involved. If they are focused on the result, things will go fine. If they are focused on doing Scrum, your project will fail because you will not get the maximum result.

My big advice: implement Scrum as a useful tool to speed up and control projects while empowering the team members, but stay focused on what it is all about: focus on the potentially shippable product we are meant to deliver!

The following has been bothering me for a while, having been closely involved in the fat client –> thing client –> rich client movements during the past decade.

Up to a few years ago everybody was talking about thin clients: zero install software doing as little as possible on the client was the future for sure. It was the future for services spanning multiple devices (such as mobile phones).

Too often I have heared the phrase: “Oh no! It will become a fat client, think about the deployment nightmare!”

These arguments seemed to be used to justify very poor HTML-based user interface for applications. And the best trick: some people kept stressing that a “website” is something completely different than an “application” - another justification for the poor (but thin!) user interfaces.

Summarized: “thin client=good” and “fat client=bad”.

Today we are in the middle of a boost in the Rich Internet Application arena: Adobe AIR, Microsoft SilverLight and Sun’s JavaFX are quickly becoming popular. In my opinion rich clients are great: they provide excellent cross-device and cross-operating system platforms (they are all pushing rollout on mobile devices).

But what is most interesting: we are now back to fat clients! Rich == Fat (ooops: “fat client=bad”).

Conceptually we are moving logic back to the client! Nobody is talking about “it is bad to move logic to the client” anymore, the discussion has moved to preventing duplicate logic. In my opinion a much healthier discussion.

The thin client obsession has been pushed too far during the end 90’s and early 2000’s: it was great to push the World Wide Web, but not so great for serious applications. RIAs can solve this now!

This weekend I was again catching up with my personal administration. Nowadays most of it involves scanning paperwork into PDFs, and typing over account numbers and Euro amounts into my online banking application.

What is amazing is that we are living in an age of web innovation, crowd sourcing, social communities, and who knows what more (but getting more advanced everyday!)… and still I have to do this prehistoric typing-over work!

So what is the problem?

In The Netherlands we have privver: a digital mailbox for receiving mail in digital form and process invoices digitally! Privver has been around for many years now but has never seen its breakthrough. The problem in my opinion: too much depending on a single company (although it is the company you would expect it from - the national mail company).

I propose an open protocol! For processing invoices and payments. And linked to traditional e-mail. Only problem is privacy sensitive information: but perhaps we should finally start using all the encryption and security options that have been out there for all those years being largely unused.

Anyway, perhaps I should give privver one more try.

There is a new, €14,95 costing, social networking hype: Poken. These are small USB-equipped characters:

As a user you can program it such that it knows your profile. When you hold it in front of another person’s poken, it will exchange the profiles of both users, confirmed by a green light. When you come home, you plug your poken into your USB port and you have new “friends” in your social network.

OK, this whole idea will become a hype. It will work in the short term. Because people like gadgets like this, it is fun and new, and it profits from the social networking hype already going on. But in the end this is another gadget that you will forget to have around at the right now. So… no, this thing is deemed to fail in the long term.
A small hint for the poken company: try to find a way to get it inside everyone’s mobile. That is a device that people already carry around all the time. How that would practically work might be a bit of an issue. But I’ll leave that up to the Poken company.
Regardless, I might buy one for the fun of it.

During the past months I have been busy with my team to arrange a new hosting location in a brand new datacenter in The Netherlands. It is great and very modern.

Perhaps old news for some, but I had not read about it yet until a few days ago: I read about this datacenter, which looks even better (actually: amazing!):

This is Pionen datacenter, located in a former nuclear bunker 30 metres below the bedrock of Stockholm in Sweden! It has 1110 square metres of space and is located within solid granite right inside the city.

And there is much more, read all about it here:

http://royal.pingdom.com/2008/11/14/the-worlds-most-super-designed-data-center-fit-for-a-james-bond-villain/

In 2008 Agile saw its true breakthrough. Especially Scrum has been benefitting.

The agile movement is a very important thing for IT: it returns focus on empowering - empower the teams and people to do their job as effective as possible and let them learn from their own mistakes.

The methodologies of the years before this have focused on preventing mistakes and enforcing a detailed process onto teams and people. This can work if the people can be effective and are empowered. But instead the traditional top-down (project) management approaches florished and the opposite happened.

Empowering is important! But let’s not forget the things we learned with previous methodologies in the past decades. There is an art to software engineering, design, etc. Let’s not forget that just empowering teams still requires expertise and know-how.

I will do everything I can to make Agile a success and promote it. One thing worries me: it is becoming a hype, and hypes lead to misunderstandings and an opposite movement (which would lead back to traditional management methods). Let’s be aware of this such that we can prevent this from happening and thus make 2009 and the years to come the years of empowerment!

Recently I was visiting the Adobe MAX conference in Milan.

I like Adobe for various reasons. But like with everything there are positive and not so positive sides to a story. Also with Adobe.

The thing is that Adobe was actually very proudly presenting a killer feature: ActionScript for your business logic on the server (using ColdFusion). In Adobe’s opinion it is a major advantage that all those front-end developers can now finally code all your vital business logic in ActionScript.
Of course, it is a major step forward compared to all the ColdFusion XML logic that Adobe has thrown in the market already.

I mostly found it funny. But I’ve been wrong in the past, and perhaps ActionScript on the server will take off anyway. It may actually replace some of the PHP and other scripted back-ends, and in that sense it is not too bad if you think about it. Also, ActionScript 3 has all the OOP elements you need and in that sense is very well maintainable (my team is proving it with a large application as we speak). So also this story has two sides.

But still: sorry Adobe.

Adobe is doing a lot of great things right now that I will blog about in the near future for sure.

Take care,
Klaas

So this week Google released its own web browser: Google Chrome.

I have been playing around with it, and am very impressed with its stability - and it performs really really well! The difference in webpage loading performance is very very noticeable and beats the experience with IE and Firefox in every way.

And the amazing thing is that it is caused by one simple fact: a fast performing Javascript engine. In a way you cannot imaging that all these years we have been browsing the web with outdated Javascript engines that could have been optimized years ago. The competition between Firefox and IE has not been sufficient to cause competition in that field, and that shows one thing: we should be very glad that there is a third very serious player in this market now.

The good news is that Firefox will soon release a similar performing javascript engine.

Final note: Google Chrome does have compatibility issues with various websites (including this WordPress installation’s admin site), so Firefox as backup browser will be my way of browsing the web for now.

Long story short, if you haven’t already tried it: Try Google Chrome!

Goed news, now there is a Dutch Flex user group. Well… good news for people in The Netherlands.

Great to see that Albumprinter made it Red Herring’s list of 200 finalists for the top 100 tech startups in Europe.

We have only recently been introducing Agile development practices in a more serious manner at Albumprinter, so I guess that has not been taken into account.